FreeTTS / News: Recent posts

It is with great pleasure that we announce the 1.2 release of FreeTTS:

http://freetts.sourceforge.net/

In this release, we have provided the following new features and
improvements over the 1.2.1 release:

- Introduced an AliasDiphone class as a subclass of Diphone, which has a name and an original diphone; most API calls are just forwarded to the original diphone
- In text db files, introduced a new type of line, "ALIAS new old", where "old" is the name of an existing diphone (must have been defined higher-up in the text db file using the traditional "DIPHONE...." mechanism), and "new" is the new name under which the diphone is to be known additionally. DiphoneUnitDatabase.parseAndAdd () has been adapted accordingly.
- In binary db files, introduced a new magic code identifying an alias diphone entry. Diphone.loadBinary() has been adapted accordingly.
- Fixed "Class-Path" issue in manifest created by FestVoxToFreeTTS.
- Fixed internal class loader problems
- using java logging library instead of system.out.println
- separation into a core jar and a jsapi1.0 dependent jar
- Bugfixes in JavaClipAudioPlayer
Catching runtime exceptions if the line is not available
writing directly to the output line instead of buffering the data
- some code cleanup... read more

The CVS repository for FreeTTS is no longer being used. It has been replaced in favor of subversion.

It is with great pleasure that we announce the 1.2 release of FreeTTS:

http://freetts.sourceforge.net/

In this release, we have provided the following new features and
improvements over the 1.2 beta2 release:

- Improvements to the emacspeak server demo
- Performance improvements for time-to-first-sample and cancel
- Improvements to the state name expansion logic (bug 1101229)
- Pronunciations for British Sterling and "not" sign (bug 1100888)
- New pronunciations for Linux terms (bug 1006630)
- Clarifications to speech.properties message (rfe 1041437 and bug 1041296)
- Fix for -dump{Multi}Audio problem in freetts app (bug 986503).
- Elimination of hang with JDK1.5 (bug 1019337)
- Tools specifically for importing CMU ARCTIC Voices.
- Fix for MBROLA phoneme remapping (bug 1013781).
- Rudimentary internationalization to FestVox import.
- Support to build FreeTTS w/o requiring JSAPI
- Only build tests on demand (more freedom from JUnit)... read more

There have been a large number of FreeTTS downloads since we released it in December of 2001. I'm personally curious about who is using our work. To help with this, and to give users a place to advertise their work, I have created a "Sightings" forum. If you're a FreeTTS user, please post a note to the "Sightings" forum to let us know more about your project:

http://sourceforge.net/forum/forum.php?forum_id=392174... read more

I'm pleased to announce the release of FreeTTS 1.2beta2. FreeTTS 1.2beta2 is mostly a maintenance release over FreeTTS 1.2beta, but includes enough changes that I don't feel comfortable releasing FreeTTS 1.2 final.

FreeTTS 1.2beta2 is available at http://freetts.sourceforge.net/, and includes the following changes over FreeTTS 1.2beta:

- Added experimental CMUArcticVoice that may (or may not) work with the CMU ARCTIC databases (http://festvox.org/cmu_arctic/). I'm looking for a collaborator to help get these voices into FreeTTS. It's not trivial and it's not for the impatient. If you're up to the challenge, drop me a line.... read more

I'm pleased to announce the release of FreeTTS 1.2beta. FreeTTS 1.2beta is feature-equivalent to Flite 1.2, including the ability to import voice data directly from FestVox voices. With this release, you can now make your voice the voice of FreeTTS! Give it a shot and be the first in your neighborhood to submit your voice to the FreeTTS effort.

In addition to being able to import voice data from FestVox, FreeTTS 1.2beta includes a lot of significant changes over FreeTTS 1.1.2:... read more

I'm pleased to announce the release of FreeTTS 1.1.2. This release contains fixes to a few pesky bugs in the FreeTTS 1.1.1 release, including the following:

- Fixed bug in TokenToWords that causes abbreviations like "OK" and "HI" to be spoken as a US state rather than just "ok" and "hi".

- Fixed race condition in JavaStreamingAudioPlayer that would cause output to not be totally cancelled.... read more

The FreeTTS project is the featured article at http://www.javasoft.com for the week of December 19.

FreeTTS is a speech synthesis system written entirely in the Java programming language.

Welcome and thank you for visiting the FreeTTS open source project. We're very excited to make a speech synthesizer written completely in the Java(tm) programming language available for free under a BSD-style license. In addition, we're also happy to make the source code available as well.

We'd like to acknowledge Alan Black and Kevin Lenzo from Carnegie Mellon University for their work with the Flite (http://www.speech.cs.cmu.edu/flite/) synthesis engine, and also acknowledge the University of Edinburgh for their work
with the Festival Speech Synthesis System
(http://www.cstr.ed.ac.uk/projects/festival/). Their expertise and hard work helped make this project succeed.... read more